OCR Text |
Show concentration of soot nuclei and their rate of growth. This effort revealed that soot precursor growth by pyrolysis is slower than oxidation up to a precursor size of pyrene (Cj6). Description of the pyrolysis and oxidation paths, identification of pyrene as the threshold size for soot growth, and observation of acetylene as growth species, and their kinetic modeling constitute the major contribution of the modeling of the Fuel Preheating Zone. Parametric calculations with this model have allowed prediction of optimum temperatures and residence times where sufficient concentrations of soot to produce a highly luminous flame are generated. Chemical kinetics and determination of species concentrations provides no information on the heat transfer characteristics of a combustion system. Energy balances must be calculated using so-called heat transfer models. Major contributions to the heat transfer of the flame system are heat transfer from the flame to the glass surface by direct radiation, indirect radiation from the refractory surfaces, and convection. Since the high temperature combustion products are a volumetrically absorbing/emitting media, the success of these heat transfer calculations depends on an accurate prediction of the spectral emission and absorption characteristics of the gaseous species (primarily C O , C O 2 and H 2 O ) at high temperature. A n efficient computational procedure for solving the Radiative Transfer Equation (RTE) is needed to calculate the complex radiative interactions among the furnace gases, refractory walls, glass batch and molten glass. Recent publications indicate that a discrete-ordinates solution of the R T E coupled with a weighted sum of gray gases ( W S G G ) model of the combustion gases can provide sufficiently accurate and efficient prediction of radiative heat transfer in this complex multicomponent system.7 The emmitance of H2O/CO2 and H20/C02/soot mixtures was predicted using different sum-of-the-gray-gas models employing a "one-dimensional" (1-D) furnace model. Fuel-soot mixture burnout was modeled parametrically, and radiation heat transfer from the flame was calculated using the total emmitance concept. Heat transfer along the flame was neglected. One of the approximations incorporated into a 1-D furnace model is that it does not account for the radiative transfer in the predominant flow direction. Gas temperature distribution along the furnace and the total heat flux along the load were calculated. In order to relax the idealization incorporated into a 1-D model and to examine the premise made, a quasi-two-dimensional (quasi 2-D) model has been developed to evaluate the importance of the axial radiative transfer for the thermal performance of the furnace. After convergence of the quasi 2-D heat transfer model was 6 |